GABAA and glycine receptor-mediated transmission in rat lamina II neurones: relevance to the analgesic actions of neuroactive steroids

Abstract

Analgesic neurosteroids such as 5alpha-pregnan-3alpha-ol-20-one (5alpha3alpha) are potent selective endogenous modulators of the GABA(A) receptor (GABA(A)R) while certain synthetic derivatives (i.e. minaxolone) additionally enhance the function of recombinant glycine receptors (GlyR). Inhibitory transmission within the superficial dorsal horn has been implicated in mediating the analgesic actions of neurosteroids. However, the relative contribution played by synaptic and extrasynaptic receptors is unknown. In this study, we have compared the actions of 5alpha3alpha and minaxolone upon inhibitory transmission mediated by both GABA(A) and strychnine-sensitive GlyRs in lamina II neurones of juvenile (P15-21) rats. At the near physiological temperature of 35 degrees C and at a holding potential of -60 mV we recorded three kinetically distinct populations of miniature IPSCs (mIPSCs): GlyR-mediated, GABA(A)R-mediated and mixed GABA(A)R-GlyR mIPSCs, arising from the corelease of both inhibitory neurotransmitters. In addition, sequential application of strychnine and bicuculline revealed a small (5.2 +/- 1.0 pA) GlyR- but not a GABA(A)R-mediated tonic conductance. 5alpha3alpha (1-10 microm) prolonged GABA(A)R and mixed mIPSCs in a concentration-dependent manner but was without effect upon GlyR mIPSCs. In contrast, minaxolone (1-10 microm) prolonged the decay of GlyR mIPSCs and, additionally, was approximately 10-fold more potent than 5alpha3alpha upon GABA(A)R mIPSCs. However, 5alpha3alpha and minaxolone (1 microm) evoked a similar bicuculline-sensitive inhibitory conductance, indicating that the extrasynaptic GABA(A)Rs do not discriminate between these two steroids. Furthermore, approximately 92% of the effect of 1 microm 5alpha3alpha upon GABAergic inhibition could be accounted for by its action upon the extrasynaptic conductance. These findings are relevant to modulation of inhibitory circuits within spinally mediated pain pathways and suggest that extrasynaptic GABA(A)Rs may represent a relevant molecular target for the analgesic actions of neurosteroids.

Figure 1

GABA_AR and GlyR mIPSCs exhibit distinct properties A, representative current traces (1.5 s) of GABA_AR and GlyR mIPSCs recorded from two exemplar LII neurones in the presence of strychnine hydrochloride (0.5 μm; top traces) and bicuculline methobromide (10 μm; bottom traces), respectively. B, representative, ensemble averages of GABA_AR and GlyR mIPSCs are shown superimposed on the same time scale to emphasize their distinct kinetic properties. C, histogram and corresponding cumulative probability plot (inset) of the decay time constants of all GABA_AR (black) and GlyR (grey) mIPSCs (n = 5054 and 6082 mIPSCs pooled from 73 and 77 cells, respectively). D and E, representative traces of individual GABA_AR (D) and GlyR mIPSCs (E) possessing mono- (left panels) or bi-exponential (right panels) decay kinetics. Curve fits (smooth black lines) are shown superimposed on the decay phase. The time constant(s) derived from exponential fitting of the decay are provided beside each mIPSC (τ: decay time constant of mono-exponential mIPSCs; τ_fast and τ_slow: fast and slow decay time constants of bi-exponential mIPSCs, respectively; A_fast: relative contribution of τ_fast to the overall peak amplitude of bi-exponential mIPSCs). Overall, GlyR mIPSCs exhibited larger peak amplitudes and faster decay kinetics than GABA_AR mIPSCs.

Figure 2

Mixed GABA_AR–GlyR mIPSCs can be distinguished from pure GABA_AR and GlyR mIPSCs A, summary of the criteria used to identify mixed mIPSCs recorded under ‘control’ conditions (i.e. in the absence of bicuculline and strychnine). B and C, data are derived from analysis of all mIPSCs recorded in the presence of strychnine (0.5 μm) or bicuculline (10 μm), i.e. pharmacologically isolated GABA_AR and GlyR mIPSCs (n = 73 and 77 cells, respectively). B, cumulative probability plot of the fast decay time constants (τ_fast) of bi-exponential GABA_AR and GlyR mIPSCs and τ-values of mono-exponential GlyR mIPSCs. Note that, due to the large overlap of these distributions (see grey shaded area), τ_fast cannot be used to distinguish between the glycinergic or GABAergic origin of bi-exponential mIPSCs. C, cumulative probability plot of the slow decay time constants (τ_slow) of bi-exponential GABA_AR and GlyR mIPSCs and τ-values of mono-exponential GABA_AR mIPSCs. Note that bi-exponential GlyR mIPSCs can be excluded by restricting the analysis to mIPSCs with a τ_slow > 10 ms (see grey shaded area). D and E, identification of bi-exponential GABA_AR and mixed mIPSCs. D, cumulative probability plot of the A_fast (%) values of mIPSCs recorded under ‘control’ conditions (i.e. in the absence of strychnine and bicuculline; in grey) and after (in black) application of 0.5 μm strychnine (n = 8 cells). Events with a τ_slow < 10 ms (i.e. putative GlyR mIPSCs) were excluded from this analysis. Note that strychnine virtually abolished (i.e. ∼90%) events with an A_fast > 50% (see grey shaded area), consistent with the removal of a fast, glycinergic component. E, scatter plot of the peak amplitude and A_fast (%) of mIPSCs recorded before (in grey) and after (in black) application of 0.5 μm strychnine (n = 8 cells). Under control conditions, a significant proportion of the mIPSCs with an A_fast > 50% also possessed relatively large peak amplitudes (> 100 pA; compare top and bottom quadrants). By contrast, strychnine-isolated GABA_AR mIPSCs were concentrated in the lower left quadrant (i.e. A_fast < 50% and peak amplitude < 100 pA). F, plot showing the relative contribution of GlyR, GABA_AR and mixed mIPSCs identified from individual cells (n = 13 cells) under control conditions by applying the criteria summarized in A. Synaptic transmission was generally clearly dominated by either pure GABA_AR or GlyR mIPSCs, although these 2 populations were more evenly represented in some cells (e.g. cell 4; see shaded box). Mixed mIPSCs represented a small proportion of the mIPSCs that were kinetically analysed (∼10%; see dotted line). G, bar chart comparing the mean peak amplitudes of GABA_AR, GlyR and mixed mIPSCs (n = 13 cells). The insets above each bar illustrate the profiles of representative, individual mIPSCs from each category (calibration: 40 pA, 10 ms). Note that the mean peak amplitude of mixed mIPSCs did not equal the sum of the individual glycinergic and GABAergic components.

Figure 3

The endogenous neurosteroid 5α3α selectively prolongs the decay phase of GABA_AR mIPSCs recorded from LII neurones A–C, cumulative probability plots of the decay time constants of all GABA_AR mIPSCs recorded before (black) and after application of 1 (A), 3 (B) and 10 μm (C) 5α3α (grey). In each of the cells shown, 5α3α produced a significant, rightward parallel shift of this relationship, indicating that all mIPSCs were sensitive to the steroid (P < 0.01; KS test). The insets show the normalized ensemble averages of GABA_AR mIPSCs recorded from the corresponding neurones before and after application of 5α3α. D, cumulative probability plot of the decay time constants of all GlyR mIPSCs recorded before (black) and after application of 10 μm 5α3α (grey). 5α3α (10 μm) did not significantly alter this distribution (P > 0.01; KS test). The inset shows the normalized ensemble averages of GlyR mIPSCs obtained from the same neurone before and after application of 5α3α. E, bar chart summarizing the effect of 5α3α on the decay kinetics (i.e. τ_cum, expressed as a percentage of control) of GABA_AR and GlyR mIPSCs. 5α3α (1–10 μm) selectively prolonged the synaptic decay of GABA_AR mIPSCs in a concentration-dependent manner. **P < 0.01; steroid versus control; 1-way RM ANOVA; †P < 0.001; GABA_AR versus GlyR; 2-way RM ANOVA; (n = 3–6 cells).

Figure 4

Mixed inhibitory synapses in LII neurones are a molecular target for 5α3α A, bar graph summarizing the effect of 1 μm 5α3α and 1 μm 5α3α+ 0.5 μm strychnine on the τ_slow, τ_w, peak amplitude and A_fast (%) of the bi-exponential mIPSCs with a putative GABAergic component (i.e. τ_slow > 10 ms). Data are expressed as percentage of control (*P < 0.05; **P < 0.01; 5α3α or 5α3α+ strychnine versus control; †P < 0.05, ††P < 0.01; 5α3αversus 5α3α+ strychnine; 1-way RM ANOVA; n = 4 cells). B, cumulative probability plots of the τ_slow values of bi-exponential mIPSCs recorded under control conditions (dotted line), in the presence of 1 μm 5α3α (grey line) and following coapplication of the steroid and 0.5 μm strychnine (black line) to cells 1 (top) and 2 (bottom). Application of 1 μm 5α3α induced a non-parallel rightward shift of the τ_slow distribution. The grey shaded area illustrates the area of overlap between the τ_slow distributions (τ_slow < 10 ms; control versus 1 μm 5α3α). Note that the mIPSCs falling in this region were insensitive to the steroid and, consistent with being purely glycinergic, were abolished upon application of strychnine. C, cumulative probability plots of the A_fast (%) values of bi-exponential mIPSCs recorded from cell 1 (top) and 2 (bottom) under each of the recording conditions described in B. Steroid-insensitive mIPSCs (i.e. τ_slow < 10 ms) were excluded from this analysis. Co-application of 1 μm 5α3α and 0.5 μm strychnine (black line) abolished nearly all mIPSCs with an A_fast > 50%, indicating that such events possessed a glycinergic component and were therefore most likely of mixed origin (see grey shaded areas of top and bottom panels). D, superimposed ensemble averages of bi-exponential mIPSCs recorded from cells 1–4 in the presence of 1 μm 5α3α (grey) and following coapplication of steroid and 0.5 μm strychnine (black). Strychnine selectively reduced the amplitude of the fast decay component in all cells but the reduction was larger for cell 1 in comparison to the other 3 cells. E, normalized ensemble averages of mixed mIPSCs recorded from cell 1 before (black) and after application of 1 μm 5α3α (grey). Note that 1 μm 5α3α selectively prolonged the slow (GABAergic) decay component of mixed mIPSCs.

Figure 5

The synthetic neuroactive steroid minaxolone prolongs the decay phase of GlyR mIPSCs from LII neurones A–C, cumulative probability plots of the decay time constants of all GlyR mIPSCs recorded from the same exemplar neurone before (black) and after application of 1 (A), 3 (B) and 10 μm (C) minaxolone (grey). Minaxolone produced a significant, rightward parallel shift of this relationship, indicating that all mIPSCs were sensitive to the steroid (P < 0.01; KS test). The insets show the corresponding normalized ensemble averages of GlyR mIPSCs recorded before and after application of minaxolone. D, bar chart summarizing the effect of minaxolone on the decay kinetics (i.e. τ_cum, expressed as a percentage of control) of GlyR mIPSCs. Minaxolone (1–10 μm) produced a small, but significant prolongation of the synaptic decay of GlyR mIPSCs. *P < 0.05; **P < 0.01; 1-way RM ANOVA; n = 3–6 cells; steroid versus control.

Figure 6

A comparison of the effect of minaxolone and 5α3α on the decay kinetics of GABA_AR mIPSCs from LII, nRT and VB neurones A, representative current traces (1.5 s sections) of GABA_AR mIPSCs recorded from exemplar LII (left), nRT (middle) and VB (right) neurones, shown on the same time scale to compare their decay kinetics. Both LII and nRT exhibited slow-decaying GABA_AR mIPSCs while the kinetics of the VB mIPSCs were ∼5-fold faster. B, representative, normalized ensemble averages of GABA_AR mIPSCs from LII (left), nRT (middle) and VB (right) neurones recorded before (black) and after (grey) application of minaxolone (300 nm). Minaxolone (300 nm) produced a clear prolongation of LII and nRT GABA_AR mIPSCs while being ineffective upon VB GABA_AR mIPSCs. C, bar charts comparing the effects of minaxolone and 5α3α on the decay kinetics (i.e. τ_cum, expressed as a percentage of control) of LII (left), nRT (middle) and VB (right) GABA_AR mIPSCs. *P < 0.05; **P < 0.01; 1-way RM ANOVA: steroid versus control; †P < 0.05: 2-way RM ANOVA: minaxolone versus 5α3α; n = 3–8 cells.

Figure 7

GABAergic and glycinergic extrasynaptic inhibition in LII neurones A and B, raw current traces (10 min) and corresponding all-points histograms illustrating the amplitude of the holding current under control conditions (2 mm kynurenic acid, 0.5 μm TTX; black) and after application of 1 μm strychnine (A) or 30 μm bicuculline (B) (grey) to two exemplar LII neurones. A small outward GlyR- but not GABA_AR-mediated tonic current was evident under these control recording conditions. C and D, plots showing the antagonist-induced changes in the RMS noise observed in the cells shown in A and B. Data points are derived from RMS measurements obtained every 51.2 ms over the 10 min recording periods. Application of either 1 μm strychnine (C) or 30 μm bicuculline (D) produced a clear and consistent decrease in the baseline (RMS) noise. E and F, raw current traces (E: 42 min; F: 14 min) and corresponding all-points histograms from two exemplar LII neurones illustrating the amplitude of the holding current under control conditions (2 mm kynurenic acid, 0.5 μm TTX, and 30 μm bicuculline (E) or 1 μm strychnine (F); in black), in the presence of either the GlyT2 inhibitor ORG25543 (100 nm; E) or the non-selective GABA transporter inhibitor nipecotic acid (1 mm; F) (in white) and following coapplication of the transporter inhibitor and corresponding GlyR and GABA_AR antagonist (in grey). Both ORG25543 and nipecotic acid induced a strychnine- and bicuculline-sensitive inward current, respectively, and increase in the baseline noise. G and H, raw current traces (23 min) and corresponding all-points histograms from two exemplar LII neurones illustrating the amplitude of the holding current under control conditions (2 mm kynurenic acid, 0.5 μm TTX, 1 μm strychnine; in black), in the presence of either 1 μm 5α3α (G) or 1 μm minaxolone (H) (in white) and following coapplication of the steroid and 30 μm bicuculline (in grey). Both steroids induced a bicuculline-sensitive inward current and increase in the baseline noise.